Toward transparent intelligence: Explainable stacked ensembles learning for LiDAR point cloud segmentation
Afridi Rahman Bondhon, Emon Kumar Dey · 2026
Read Toward transparent intelligence: Explainable stacked ensembles learning for LiDAR point cloud segmentation by Afridi Rahman Bondhon, Emon Kumar Dey (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/plos-pone-0345125.txt.
Description
Afridi Rahman Bondhon and Emon Kumar Dey built an explainable system for labeling three-dimensional laser scans of terrain and cities, point by point. Stacking several classifiers usually lifts accuracy but buries the reasoning. Their framework stacks multiple base learners under a simple final model and then applies explanation tools to show which features drove each decision.
On two benchmark datasets the stack labeled points with accuracy above ninety percent. The explanations also revealed which input features could be dropped, cutting training time by at least seven percent without losing accuracy. For city planning and environmental monitoring, a segmentation tool that can justify itself is easier to audit and trust than a slightly stronger black box.
Catalog record
| CREATOR | Afridi Rahman Bondhon, Emon Kumar Dey |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 71976 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |